There are couple of functions allowing to estimate panel fixed effects model, those include xtreg, areg, reghdfe. The areg allows for fixed weights weighted fixed effects panel regression. The reghdfe allows for time varying weights in fixed effects panel regression. The xtreg can give random effects model estimates. Is there a command that would allow for time varying weights random effects weighted regression?
A concrete example would be to estimate some population effect between variables where a sample coming from some strata is of different size than the strata in the population, for example data coming from european countries where countries have their number of people in each year and the sample needs weighting to account for nationality sizes, because the counties sample share is not proportional to their nationality sizes. This should be a standard problem in survey and census data shouldn't it? If there is not a simple command that does so is there a recipe for it or materials to build own function?
A concrete example would be to estimate some population effect between variables where a sample coming from some strata is of different size than the strata in the population, for example data coming from european countries where countries have their number of people in each year and the sample needs weighting to account for nationality sizes, because the counties sample share is not proportional to their nationality sizes. This should be a standard problem in survey and census data shouldn't it? If there is not a simple command that does so is there a recipe for it or materials to build own function?
